Short Bio: Animesh Garg is the Stephen Fleming Early Career Professor in Computer Science at Georgia Tech, within the School of Interactive Computing, and is affiliated with the Robotics and Machine Learning programs. He holds courtesy appointments at the University of Toronto and the Vector Institute. Previously, he held research leadership positions at Nvidia and Apptronik. His research focuses on the algorithmic foundations of generalizable autonomy, enabling robots to acquire cognitive and dexterous skills and collaborate with humans in novel environments. His group explores structured inductive biases, causality in decision-making, multimodal object-centric representations, self-supervised learning for control, and efficient dexterous skill acquisition.

Short Bio: Daniel Ho is the Director of Evaluation at 1X Technologies. His goal is to deploy generalist machines that grow from experience and correct their own mistakes. He's building World Models and large-scale evaluation pipelines towards this mission. Previously, he worked on robotics, perception, and machine learning as a Senior Software Engineer at Waymo and Everyday Robots (X, Google[X]). His research has focused on learning algorithms and representation learning to generalize ML model understanding in robotics, computer vision, and self-driving.

Talk Title: 1X World Model: Solving humanoid policy training and evaluation with data synthesis and action control

Short Bio: Hao Su is an Associate Professor of Computer Science at UC San Diego and Founder & CTO of Hillbot, a robotics startup. He directs the Embodied Intelligence Lab and is a founding member of the Halıcıoğlu Data Science Institute. His research spans computer vision, machine learning, graphics, and robotics, focusing on algorithms to simulate and interact with the physical world. He holds Ph.D.s in Computer Science from Stanford and Mathematics from Beihang University. He helped develop datasets like ImageNet, ShapeNet, and tools like PointNet. Su is Program Chair of CVPR 2025 and has received NSF CAREER and SIGGRAPH awards.

Talk Title:Learning World Models for Embodied AI

Short Bio: Katerina Fragkiadaki is the JPMorgan Chase Associate Professor in Machine Learning at Carnegie Mellon University. She earned her B.S. from the National Technical University of Athens and her Ph.D. from the University of Pennsylvania, followed by postdoctoral work at UC Berkeley and Google Research. Her research combines common sense reasoning with deep visuomotor learning to enable few-shot and continual learning for perception, action, and language grounding. Her group develops methods in 2D-to-3D perception, vision-language grounding, and navigation policies. She received awards including the NSF CAREER and DARPA Young Investigator Awards and is Program Chair for ICLR 2024.

Talk Title: From Videos to Physics Engine Simulations to Neural Simulations

Short Bio: Yilun Du is an Assistant Professor at Harvard’s Kempner Institute and Computer Science Department, and a Senior Research Scientist at Google DeepMind. He earned his Ph.D. in EECS from MIT, advised by Leslie Kaelbling, Tomas Lozano-Perez, and Joshua Tenenbaum. He holds a bachelor’s from MIT and has been a research fellow at OpenAI and a visiting researcher at FAIR and DeepMind. A gold medalist at the International Biology Olympiad, his research focuses on generative models, decision making, robot learning, and embodied agents. He develops energy-based models enabling generalization and advances in diffusion models, scene understanding, and trajectory planning.

Talk Title: Learning Generative World Simulators

Short Bio: Nicholas Pfaff is a third-year PhD student in the Robot Locomotion Group at MIT, advised by Prof. Russ Tedrake. His research focuses on scaling realistic simulations to generate training data for robotic manipulation foundation models, combining model-based and learning-based approaches. Prior to MIT, he worked at Ocado Technology on large-scale pick-and-place systems, and he is a recipient of an MIT Graduate Teaching Award.

Talk Title: Scaling Robot Simulation: Automating Physically Accurate and Diverse Worlds

Short Bio: Boyi Li is a Research Staff at NVIDIA Research and a Research Fellow at UC Berkeley. She earned her Ph.D. in EECS from Cornell University & Cornell Tech. Her research focuses on advancing multimodal embodied intelligence, developing generalizable algorithms, and creating interactive intelligent systems. Central to this work is reasoning, language models, generative models, and robotics. A key aspect involves aligning representations from diverse multimodal data, including 2D pixels, 3D geometry, language, and audio.

Talk Title:Learning to Simulate Multimodal Robot Worlds